topic Re: Random Slope parameters per individual using proc mixed in Statistical Procedures

Random Slope parameters per individual using proc mixed

lgutm18 — Thu, 17 Apr 2025 15:29:52 GMT

Hello,

I need some help obtaining the random slope estimates per individual, so that I can use those parameters per individual as an outcome in another part of the analysis.

Below is the code that I am currently using:

ods output solutionf=slopes;
proc mixed data=trial;
class idno time;
model GLOBALCOG = time WMFA WMH ICV brainvol AGE RACE GENDER/s chisq OUTPM=WORK.PM RESIDUAL;
random intercept time;
repeated time/type=un subject=idno r rcorr;
run;

The part "solutionf=slopes" provide one parameter of the random slope but not per individual.

Thanks!

Re: Random Slope parameters per individual using proc mixed

PaigeMiller — Thu, 17 Apr 2025 18:05:13 GMT

Is this what you are looking for?
https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.3/statug/statug_mixed_examples05.htm

Re: Random Slope parameters per individual using proc mixed

lgutm18 — Thu, 17 Apr 2025 21:52:01 GMT

Yes! Thanks for sharing. However, what it's still not clear is how to extract the random slope parameters per individual as a column that you could add to the dataset to use for analysis. I'd greatly appreciate suggestions!

Re: Random Slope parameters per individual using proc mixed

Ksharp — Fri, 18 Apr 2025 01:10:07 GMT

I think your code should look like this:

data trial;
call streaminit(123);
do idno=1 to 10;
 do time=1 to 4;
  globalcog=rand('normal');
  brain=rand('normal');
  output;
 end;
end;
run;

ods  output SolutionR= SolutionR;
proc mixed data=trial;
class idno time;
model GLOBALCOG = time brain/s chisq  RESIDUAL;
random intercept /subject=idno solution ;
repeated time/type=un subject=idno r rcorr;
run;

Re: Random Slope parameters per individual using proc mixed

Ksharp — Fri, 18 Apr 2025 01:33:37 GMT

If this random slope is one slope for one subject. Try this:

data trial;
call streaminit(123);
do idno=1 to 10;
 do time=1 to 4;
  globalcog=rand('normal');
  brain=rand('normal');
  output;
 end;
end;
run;

ods  output SolutionR= SolutionR;
proc mixed data=trial;
class idno time ;
model GLOBALCOG = time brain/s chisq  RESIDUAL;
/*random intercept /subject=idno solution ;*/
random  brain/subject=idno solution ;

/*repeated time/type=un subject=idno r rcorr;*/
run;

Re: Random Slope parameters per individual using proc mixed

jiltao — Fri, 18 Apr 2025 13:43:31 GMT

First, your PROC MIXED code does not seem to be appropriate. It looks to me that you wanted to fit a random coefficients model? If so, TIME should not be a CLASS variable, and you should have the SUBJECT= option in the RANDOM statement. Then with this modified RANDOM statement, your REPEATED statement with TYPE=UN structure is likely over-fitting your data and might need to be removed.

proc mixed data=trial;
class idno ;
model GLOBALCOG = time WMFA WMH ICV brainvol AGE RACE GENDER/s chisq OUTPM=WORK.PM RESIDUAL;
random intercept time / subject=idno;
run;

Then you might use approaches illustrated in the following usage note to obtain subject-specific intercept and slope estimates.

37109 - Obtaining subject-specific parameter estimates and tests for a random coefficients model or HLM with PROC MIXED or PROC GLIMMIX

Hope this helps,

Jill

Re: Random Slope parameters per individual using proc mixed

lgutm18 — Sat, 26 Apr 2025 00:46:20 GMT

This works great! Thank you very much!

Re: Random Slope parameters per individual using proc mixed

lgutm18 — Sat, 26 Apr 2025 00:46:56 GMT

I appreciate your comment and the resource. I have fixed my code.
Thanks!!